AI & ML
What Survives the Shift from Text to Pixels? Conditional Protection against Prompt Injection in Image-Rendered Documents
arXiv cs.CR0 views
arXiv:2611.24231v1 Announce Type: new Abstract: Multimodal language models increasingly answer questions over external documents, exposing them to indirect prompt injection: commands planted in untrusted content to hijack the response. Rendering untrusted content as page images lowers attack success on average, leaving open which injected operations survive rendering, at what utility cost, and whether failed image attacks reflect unread commands or commands read but not followed. We pair text and page-image versions of 191 English question-answering sources across six models, measuring marker appending, answer formatting, and answer replacement alongside clean Question-Answering utility. Rendering reduces append and format execution for every model, yet Qwen3-VL-8B still executes image-borne replacement at 77.96\%. Protection also depends on target type, rendering settings, wording, and repetition, and a search over eight fixed attack candidates largely closes the image deficit in appending and formatting for Qwen and Gemini, though not for GPT-5-nano. Utility ratios on original questions range from 93.31\% to 101.61\%, but Qwen falls to 61.99\% when answers combine two evidence spans. Illegibility cannot explain all failed attacks: models often transcribe and identify image-borne commands they do not execute, and trusted user authorization raises execution on unchanged pages. Inside Qwen, patching early attack-region states induces all three operations, more weakly in images. Attention blocking suggests why: a text command acts through its own positions, whereas visual encoding spreads an image command and its target across the instruction page. Rendering thus offers conditional protection, motivating evaluations that report security by model, operation, and attack budget alongside utility.
Source:https://arxiv.org/abs/2611.24231PreviousErased but Recoverable: On the Fragility of Concept Erasure Under Fine-TuningNextPMKPIR: Parallel Multi-Keyword PIR with Interleaved Batching for Private Boolean Retrieval
Related posts
Does your AI agent need an LLM to decide what to do next?
0The New Stack
Expert comment: Why do so many AI tools end up in the AI graveyard?
0Phys.org Tech
Token fraud is here and it’s BAD (YouTube)
0Simon Willison

LiberNovo’s Black Friday sale takes up to 42% off ergonomic chairs
09to5Mac

WhatsApp is raising group chat limit to 1,500 participants
09to5Mac
Who called me about stolen ID and loans? Handle it like an unauthenticated security alert
0dev.to